AIPython★ 248⑂ 23 forks
yuanzhao-CVLAB/UniMMAD
UNIMMAD ANOMALY DETECTION IN MULTIMEDIA MODEL
UniMMAD offers a unified approach to anomaly detection in multimedia data by combining images, audio, and text. Using mixture‑of‑experts, it performs feature decompression, highlighting useful signals and suppressing noise.
// KEY FEATURES
- Combines images, audio and text in a single multimodal pipeline for an joint analysis.
- Applies Mixture‑of‑Experts for selecting specialized subnets for each specific data type.
- Performs feature decompression via expert modules removing noise and amplifying signals
- Achieves SOTA on real-world benchmarks for multiclass anomaly detection
New repositories every 30 minutes
REDDYX AI scans GitHub 24/7 and ships the best AI/ML/Web3 projects to Telegram.
Join on Telegram